I have a chunk of data which is supposed to be zlib compressed data (I was not 100% sure).
I first tried to uncompress it with gzip by prepending "1F 8B 08 00 00 00 00 00". Just like in the accepted answer of this thread (https://unix.stackexchange.com/questions/22834/how-to-uncompress-zlib-data-in-unix). It worked out and it was probably the right approach, because the output contained a lot of human readable strings.
I then tried to implement this in a c++ program using zlib. But it seems that zlib generates a different output. Am I missing something? zlib and gzip should be basically the same (despite the headers and trailers), shouldn't they? Or do I have a simple error in my code below? (the chunk of data is shortened for the sake of simplicity)
unsigned char* decompressed;
unsigned char* dataChunk = /*...*/;
printHex(dataChunk, 160);
int error = inflateZlib(dataChunk, 160, decompressed, 1000);
printHex(decompressed, 160);
//zerr(error);
printHex(unsigned char* data, size_t n)
{
for(size_t i = 0; i < n; i++)
{
std::cout << std::hex << (uint16_t)data[i] << " ";
}
std::cout << std::dec << "\n-\n";
}
int inflateZlib(unsigned char* data, size_t length, unsigned char* decompressed, size_t maxDecompressed)
{
decompressed = new unsigned char[maxDecompressed];
z_stream infstream;
infstream.zalloc = Z_NULL;
infstream.zfree = Z_NULL;
infstream.opaque = Z_NULL;
infstream.avail_in = (uInt)(length); // size of input
infstream.next_in = (Bytef *)data; // input char array
infstream.avail_out = (uInt)maxDecompressed; // size of output
infstream.next_out = (Bytef *)decompressed; // output char array
// the actual DE-compression work.
int ret = inflateInit(&infstream);
zerr(ret);
ret = inflate(&infstream, Z_NO_FLUSH);
zerr(ret);
inflateEnd(&infstream);
return ret;
}
This produces the following output:
78 9c bd 58 4b 88 23 45 18 ee 3c 67 e3 24 93 cc ae 8a f8 42 10 c4 cb 1a 33 a3 7b f0 60 e6 e0 e6 e0 49 90 bd 29 4d 4d 77 25 dd 99 ee ea de aa ee 4c 32 82 2c e8 c1 93 ac 47 c5 45 f 82 8 5e 16 f ba 78 18 45 d0 83 7 95 15 5c d0 c3 aa b0 b2 ee 65 5c f0 e4 c5 bf aa 1f a9 ea 74 cf 64 7 31 c3 24 9d fa fe bf ea ab ff 59 15 ab 62 6a b5 5d 9b 8c 18 2a 5b 15 47 d3 b4 92 55 35 b5 ba b7 3d c6 46 b0 a3 35 3 1c 50 64 61 93 7a a4 67 d5 0 e1 c2 d8 e4 92 75 fe 56 b3 ca a6 76 c2 f0 1c 8f
-
0 0 6 c0 83 50 0 0 16 b0 78 9c bd 58 4b 88 23 45 18 ee 3c 67 e3 24 93 cc ae 8a f8 42 10 c4 cb 1a 33 a3 7b f0 60 e6 e0 e6 e0 49 90 bd 29 4d 4d 77 25 dd 99 ee ea de aa ee 4c 32 82 2c e8 c1 93 ac 47 c5 45 f 82 8 5e 16 f ba 78 18 45 d0 83 7 95 15 5c d0 c3 aa b0 b2 ee 65 5c f0 e4 c5 bf aa 1f a9 ea 74 cf 64 7 31 c3 24 9d fa fe bf ea ab ff 59 15 ab 62 6a b5 5d 9b 8c 18 2a 5b 15 47 d3 b4 92 55 35 b5 ba b7 3d c6 46 b0 a3 35 3 1c 50 64 61 93 7a a4 67 d5 0 e1 c2 d8 e4 92 75
-
which is not what I want. Whereas gzip:
printf "\x1f\x8b\x08\x00\x00\x00\x00\x00\x78\x9c\xbd\x58\x4b\x88\x23\x45\x18\xee\x3c\x67\xe3\x24\x93\xcc\xae\x8a\xf8\x42\x10\xc4\xcb\x1a\x33\xa3\x7b\xf0\x60\xe6\xe0\xe6\xe0\x49\x90\xbd\x29\x4d\x4d\x77\x25\xdd\x99\xee\xea\xde\xaa\xee\x4c\x32\x82\x2c\xe8\xc1\x93\xac\x47\xc5\x45\xf\x82\x8\x5e\x16\xf\xba\x78\x18\x45\xd0\x83\x7\x95\x15\x5c\xd0\xc3\xaa\xb0\xb2\xee\x65\x5c\xf0\xe4\xc5\xbf\xaa\x1f\xa9\xea\x74\xcf\x64\x7\x31\xc3\x24\x9d\xfa\xfe\xbf\xea\xab\xff\x59\x15\xab\x62\x6a\xb5\x5d\x9b\x8c\x18\x2a\x5b\x15\x47\xd3\xb4\x92\x55\x35\xb5\xba\xb7\x3d\xc6\x46\xb0\xa3\x35\x3\x1c\x50\x64\x61\x93\x7a\xa4\x67\xd5\x0\xe1\xc2\xd8\xe4\x92\x75\xfe\x56\xb3\xca\xa6\x76\xc2\xf0\x1c\x8f" | gzip -dc | hexdump -C
produces:
gzip: stdin: unexpected end of file
00000000 68 03 64 00 05 77 69 6e 67 73 61 02 68 03 6c 00 |h.d..wingsa.h.l.|
00000010 00 00 01 68 04 64 00 06 6f 62 6a 65 63 74 6b 00 |...h.d..objectk.|
00000020 0c 74 65 74 72 61 68 65 64 72 6f 6e 31 68 05 64 |.tetrahedron1h.d|
00000030 00 06 77 69 6e 67 65 64 6c 00 00 00 06 6c 00 00 |..wingedl....l..|
00000040 00 05 68 02 64 00 08 63 6f 6c 6f |..h.d..colo|
0000004b
which is what I want.